From Crisis to Revolution: Leveraging Heterogeneity in Consumer Research for Generalizability

Some consumer research aims to affect marketing practice with rapidly applied insights. To do this, consumer research findings need to be generalized from the settings of the original research to the application. However, research results reflect both the manipulation of interest and a myriad of unobserved sources of heterogeneity that affect whether and how strong an effect occurs. To generalize, we need to understand this heterogeneity. We propose a toolbox supported by the re-analysis of existing data and simulations to help consumer researchers leverage heterogeneity for generalization. We propose five levers that consumer researchers can use to increase the generalizability: 1) Measuring proximal moderators that describe respondents' interaction with the setting. 2) Exploiting purposive variation to increase the range of observed moderators and settings. 3) Measuring manipulation intensity and measurement error. 4) Using survey para-data to estimate moderators and 5) Harnessing proximal moderators and purposive variation to predict generalized effect sizes. We suggest that our toolbox can help advance our field towards higher practical impact by moving beyond understanding what works to understanding what works when, where, and why.